The goal of the proposed research is to develop a comprehensive conceptual framework mapping motor impairments into biomechanical functional limitations and task performance to show how mobility is compromised in persons with hemiparesis post-stroke in order to improve clinical measurements of locomotor ability and functional outcomes. The framework causally associates neural output (muscle activation) with multi-muscle force production and its consequent biomechanical action on the musculoskeletal system during execution of natural locomotor tasks. Basic to this neuromechanical framework is a modular organization of muscle activation where each module (set of muscles) produces a specific biomechanical locomotor function. The modular structure and its control of muscle activation can account for not only the complex muscle activity observed during impaired and non-impaired walking but also the impaired locomotor performance caused by the combined biomechanical output from each module. Within this modular-based, neuromechanical framework we will find the changes in the number of modules (locomotor output complexity), and in modular composition (relative weighing of muscles in a module) and control (magnitude and timing), when healthy and hemiparetic subjects with residual walking impairments perform adaptive locomotor tasks. Aim 1 investigates steady-state adaptive tasks that assess a person's biomechanical reserve (maximal cadence, step length, step height and speed). Aim 2 investigates non-steady state adaptive tasks important to household/community walking (start, stop, turn, accelerate and decelerate). Analysis of the experimental data will identify how locomotor output complexity is reduced and how modular composition and control changes limit performance in subjects with different levels of motor impairment. Computer simulations using this modular structure and control of muscle activation will be generated to emulate the observed kinematics, ground reaction forces and EMGs of these adaptability tasks when performed by the healthy and impaired subjects. Simulation analyses will identify how reduced locomotor output complexity and changes in modular composition and control affect independent control and execution of the critical biomechanical locomotor functions (forward propulsion, body support, leg swing, and muscle mechanical work) and the consequent effect on locomotor-task performance. Our total integrated approach, then, relates assessments across the domains of impairment (modular changes), functional limitation (biomechanical function execution), and participation restriction (household/community walking) to provide clinicians with knowledge to better diagnose, design task-specific interventions, plan treatment, and monitor treatment effects. PUBLIC HEALTH RELEVANCE: Walking is important to persons who have had a stroke and better rehabilitation methods are needed to restore or improve their walking. This project will find new and better ways to measure walking performance that are explicitly linked to specific underlying impairments. Clinicians, such as physicians and physical therapists, will then be able to make these measurements in their clinic to better diagnose a person's specific walking deficit, design a specific treatment plan, and monitor its ability to restore or improve the person's walking.